Acrocephalus 112.qxd 16. 12. 02 15:29 Page 75 O Acrocephalus 23 (112): 75 – 79, 2002 Effect of floods on the distribution of meadow birds on Ljubljansko barje Vpliv poplav na gnezditveno razširjenost travniških ptic na Ljubljanskem barju Davorin Tome National Institute of Biology, Ve~na pot 111, SI-1000 Ljubljana, Slovenija, e-mail: davorin.tome@uni-lj.si I investigated the density of meadow birds breeding on Ljubljansko barje as a function of three different flooding regimes. The densities of Common Quail Coturnix coturnix, Corn Crake Crex crex, Northern Lapwing Vanellus vanellus, Sky Lark Alauda arvensis, Tree Pipit Anthus trivialis, Whinchat Saxicola rubetra, Marsh Warbler Acrocephalus palustris, Common Whitethroat Sylvia communis and Corn Bunting Miliaria calandra were significantly higher on regularly flooded than on non-flooded areas. A similar effect was not observed on a Common Stonechat Saxicola torquata population. I discuss some reasons for the positive effect of floods on breeding density. Preservation, or even expansion, of areas with extensive agriculture and regular floods is strongly recommend as an essential conservation measure for meadow birds on Ljubljansko barje. Key words: meadow birds, breeding density, floods, Ljubljansko barje, Slovenia Ključne besede: travni{ke ptice, gnezditvena gostota, poplave, Ljubljansko barje, Slovenija 1. Introduction In Europe, agriculture is one of the most important factors affecting bird populations (Tucker & Evans 1997). According to British experience, of all birds, those from farmland suffered the greatest decline in term of population number and distribution (Gibbons et al. 1993, Fuller et al. 1995). About 70% of 173 priority bird species from agricultural and grassland habitats in Europe have an unfavourable conservation status (Tucker & Dixon in: Tucker & Evans 1997). A similar situation pertains in North America (Herkert & Knopf 1998). It is not easy to define the key causes for these declines, since there are many but what they all have in common is, what we know as agricultural intensification. Some of the well known causes are increased use of pesticides and fertilisers, creation of large fields without set asides (hedges, etc.), changes in ploughing, harvesting, mowing regime and extensive land drainage (Newton 1998), which are often basic to all other causes considered. In Slovenia there is no hard data on temporal changes in birds populations, but some reports indicate a decline of farmland birds and birds from wet grasslands (Bra~ko 1986, Tome 1998). There are also some studies where spatial changes in breeding bird densities, caused by different types and levels of land practice, could be used to estimate the influence of agriculture on bird communities. One such is the breeding bird atlas of Ljubljansko barje (Sovinc et al in prep.). The aim of this paper is to present differences in breeding densities of meadow birds on flooded and non-flooded areas of Ljubljansko barje and to evaluate them from the land management point of view. 2. Study area and methods Ljubljansko barje is 5-10 km wide and about 20 km long, a flat depression south of the city of Ljubljana in central Slovenia. The study area is about 150 km2, and lies 287-290 m above sea level. Before agriculture started to become intensified 150 years ago, this was a raised bog. Today, primarily due to drainage and exploitation of peat, only a few raised bog fragments remain (total area about 0.5%), all of which have reached full development in forests of Pino sylvestris-Betuletum and Betulo-Quercetum roboris type (Martin~i~ 1987). The remaining areas are 75 O Acrocephalus 112.qxd 16. 12. 02 15:29 Page 76 D. Tome: Effect of floods on the distribution of meadow birds on Ljubljansko barje predominantly corn fields (about 25%), semi-intensive grasslands (about 40%), traditional meadows (about 12%), built-up and industrial areas (about 7%). Other types of land (orchards, ditches area, reed beds, open water, etc.) do not account for more than 1% of the area. (Kotarac 1999) In spite of long term efforts at drainage, occasional floods still occur. They are confined roughly to the central part of Ljubljansko barje, on both sides of the Ljubljanica river (Kolbezen 1984). Floods are most frequent in the autumn and winter, less so in spring and least of all in summer. They usually last from 1 to 7 days. Data on numbers of breeding bird populations were collected by several field workers between 1989 and 1996 during a systematic survey of the area for an atlas of breeding birds in a 1x1 km grid (detailed description of data collection methods see in Sovinc et äl. 1993). In the present work I took into consideration only squares with at least 75% non-forested, non built-up area. These squares were designated as open squares”. According to data from Kolbezen (1984), selected open squares were separated into three types: (1) regularly flooded squares - entirely flooded almost every year; (2) occasionally flooded squares - flooded only partially and, in some years, not flooded at all; (3) non-flooded squares - normally not flooded. Meadow birds were defined as all non-aquatic species nesting directly on the ground among herbaceous plants and as birds building nests low over the ground on annual plants. Densities of birds were compared between all three types of squares, using Kruskal-Wallis ANOVA test. Differences were significant to p<0.05. The calculations were done only for birds found on at least 50 open squares, to minimise inaccuracy caused by small sample size. In 1997, all squares were surveyed in summer and scored visually to the nearest 10% for the presence of forest, fields, meadows and, to the nearest 100 m length, of hedgerows ( mejice” in Slovene language). 3. Results Of 111 squares considered in this research, 47 (42%) were regarded as non-flooded, 24 (22%) as occasionally-flooded and 40 (36%) as regularly flooded, indicating that a little over half the open area on Ljubljansko barje is, at least occasionally, exposed to high waters. Table 1: Proportion of occupied squares and density of singing males on nonflooded (NF) areas, semiflooded (SF) areas and flooded (FL) areas Tabela 1: Dele` zasedenih kvadratov in gostota pojo~ih samcev v nepoplavljenih (NF), delno poplavljenih (SF) in poplavljenih (FL) obmo~jih Occupied squares/ Density of singing males/km2 (avg, std), number of Kruskal-Wallis Zasedeni kvadrati occupied squares (N) / Gostota pojo~ih samcev/km2 ANOVA (avg, std), {tevilo zasednih kvadratov (N) NF SF FL % % % Coturnix coturnix Crex crex Vanellus vanellus Alauda arvensis Anthus trivialis Saxicola rubetra Saxicola torquata Acrocephalus palustris Sylvia communis Miliaria calandra 53 23 28 77 79 81 72 77 87 30 63 67 50 100 100 96 88 100 100 63 78 83 éo IOO IOO IOO 70 IOO IOO 50 NF avg std N 1.6 2.56 25 0.3 0.60 11 2.0 4.45 13 8.4 8.49 36 9.9 7.64 37 9.4 7.95 38 2.5 2.45 34 9.2 9.36 36 4.9 4.12 41 0.4 0.69 14 avg SF std N avg FL std N H p 2.9 1.2 4.4 16.3 14.8 17.8 2.4 14.2 8.2 1.3 3.42 15 i.io 16 5.72 12 11.57 24 6.15 24 8.63 23 1.82 21 9.03 24 5.64 24 1.71 15 4.2 4.6 3.5 16.4 22.8 22.8 1.5 21.9 12.3 1.7 4.87 31 3.99 33 4.21 24 13.09 40 10.52 40 10.21 40 1.54 28 12.42 40 6.73 40 2.53 20 9.9 42.0 11.2 17.2 32.7 34.5 3.1 32.0 34.1 8.3 0.00720 0.00000 0.00360 0.00020 0.00000 0.00000 0.21050 0.00000 0.00000 0.01600 No. of squares/ Število kvadratov 47 24 40 76 Acrocephalus 112.qxd 16. 12. 02 15:29 Page 77 AcROCEPHALUs 23 (112): 75 — 79, 2002 Among meadow birds, 10 species breed on Ljubljansko barje in a defined distribution threshold (over 50 occupied squares). They are as follows (with number of occupied squares and estimated mean population size in brackets; Sovinc et al. in prep.): Common Quail Coturnix coturnix (79, 330), Corn Crake Crex crex (73, 260), Northern Lapwing Vanellus vanellus (52, 340), Sky Lark Alauda arvensis (113, 1480), Tree Pipit Ant bus trivialis (127, 1970), Whinchat Saxicola rubetra (123, 1860), Common Stonechat Saxicola torquata (102, 290), Marsh Warbler Acrocephalus palustris (127, 2060), Common Whitethroat Sylvia communis (129, 1030) and Corn Bunting Miliaria calandra (53, 125). In all species except the Common Stonechat, there were significantly different densities between non, occasionally and regularly flooded squares (Table 1). Population sizes of Common Quail, Corn Crake, Tree Pipit, Whinchat, Marsh Warbler, Common Whitethroat and Corn Bunting increased more or less steadily from non, through occasionally, to regularly-flooded areas, while populations of Northern Lapwing and Sky Lark increased significantly from non to occasionally flooded areas, but remained unchanged on regularly flooded areas. The proportion of fields and meadows correlated significantly with the flooding regime, with fields occupying almost twice as large an area in non-flooded as in flooded squares (H=9.74, p=0.0077), and a considerably higher proportion of meadows on flooded as opposed to non-flooded areas (H=10.27, p=0.006; Figure 1). The proportion of forests (H=0.18, p=NS) and length of the tree lines (H=3.77, p=NS) did not differ between areas with different flooding regimes, probably due to eliminating squares with more than 1/4 of the area covered with trees. 4. Discussion Common Stonechat was the only one of ten species with a lower density on flooded compare to non-flooded areas, which came as no surprise. In Europe, dry plains and hillsides are the most frequent type of breeding habitat of this species (Cramp 1998). Apart from this, the results indicate the great importance of floods for meadow birds (Table 1) - nine out of ten species nested in greater densities on flooded than on non-flooded areas. More than 70% of their total population, and over 90% of all calling male Corn Crakes being present on regularly and occasionally flooded areas combined (57% of the open squares). If breeding success would be considered, the value of regularly flooded, traditional meadows for birds would probably be even greater. Population sizes of the nine meadow species ranged from 10% (Tree Pipit) to about 50% (Corn Crake, and Marsh Warbler) of the total Slovenian population (Trontelj 2001, Sovinc et al. in prep.). Considering that Ljubljansko barje is less than 1% of the total area of Slovenia, the importance for birds of floods must be considered on the national scale. Four of the species, Common Quail, Corn Crake, Sky Lark and %% 90 80 70 60 50 40 30 20 10 0 60 50 40 30 20 10 0 NF SF meadows / travniki FL NF SF fields / njive FL Figure 1: Proportion of meadows and fields according to flooding regime in 1 x 1 km squares on Ljubljansko barje (NF=non-flooded, SF=semi-flooded, FL=flooded, bars=average, lines = standard deviation) Slika 1: Delež travnikov in njiv glede na poplavni režim v 1 x 1 km kvadratih na Ljubljanskem barju (NF=nepoplavni, SF=delno poplavni, FL=poplavni, bars=povprečje, črte=standardni odklon) 77 Acrocephalus 112.qxd 16. 12. 02 15:29 Page 78 D. Tome: Effect of floods on the distribution of meadow birds on Ljubljansko barje Corn Bunting are also listed as endangered breeding birds, are more common in non-flooded areas (Figure birds (Bra~ko et al. 1994), with Northern Lapwing, 1), resulting in low breeding densities. On the other Whinchat and Common Whitethroat being on a new hand, high water table and frequent floods make a proposal of endangered breeding birds in Slovenia structure of vegetation more suitable for many of the (DOPPS unpubl.). meadow birds, and moist land also increases the The most striking change in density was in Corn quantity and/or quality of available food (Beintema Crake, a globally endangered species (Tucker & 1988). So floods on Ljubljansko barje have a direct Heath 1994). On frequently flooded areas there was impact on agriculture practice and on some important about 10 times greater density than on non-flooded ecological features of the meadows, all in turn ones. These results confirm findings (Grobelnik influencing the distribution and population size of the 2000) that, on Ljubljansko barje, traditional meadows birds. If the nationally important breeding are by far the most important habitat type for this populations of meadow birds on Ljubljansko barje are species, and agree with conclusions of Willi (1985), to be conserved, it is of vital importance to preserve or that Corn Crake is among the most sensitive birds even expand areas with traditional agriculture and when drainage of the habitat is in question. Although regular floods. the population size of this species is not significant on the international scale, representing less than 0.01%, 5. Povzetek it is one of the greatest in southern Europe (Schäffer & Green 2000) – and, we should not forget, Na Ljubljanskem barju sem raziskoval gostoto concentrated on a very small area! travni{kih gnezdilcev na treh obmo~jih, ki se med Northern Lapwings and Sky Larks differ from seboj razlikujejo glede na re`im poplavljanja. Ugotovil other species whose populations increase more or less sem, da so gostote prepelice Coturnix coturnix, kosca steadily from non, through occasionally, to regularly Crex crex, pribe Vanellus vanellus, poljskega {krjanca flooded areas, in having average densities that are Alauda arvensis, drevesne cipe Anthus trivialis, similar on occasionally and regularly flooded squares. repalj{~ice Saxicola rubetra, mo~virske trstnice In spite of differences in ecology of these species, they Acrocephalus palustris, rjave penice Sylvia communis in both prefer to nest in short vegetation or, if not velikega strnada Miliaria calandra zna~ilno ve~je na available, on fields (Willi 1985, Beintema & vsakoletno poplavljenih povr{inah kot na povr{inah, Muskens 1987, Chamberlain & Gregory 1999). On kjer poplav ni ali pa so zelo redke. Gnezditvena Ljubljansko barje many regularly flooded areas stay gostota prosnika Saxicola torquata ni bila odvisna od unmown (personal observations). In spring they are poplavnega re`ima. Predstavljam nekaj vzrokov za covered with a dense, tall layer of dead annual tako ugoden vpliv poplav na ptice. Kot nujen vegetation stalks (predominantly Filipendula naravovarstveni ukrep za ohranitev nacionalno ulmarid), which makes them less appropriate for the pomembnih populacij travni{kih ptic na barju two species, hence reducing their average density in predlagam ohranitev ali celo raz{iritev poplavnih flooded squares. Sky Larks also prefer dry to wet povr{in in ekstenzivnega kmetovanja. nesting places (Willi 1985). My estimate is, that the importance of floods for these two species is to some 6. References degree lower (but not insignificant!) than for the other seven. Beintema A.J. & G.J.D.M. Muskens (1987): Nesting Since all the birds considered are terrestrial, water success of birds breeding in Dutch agricultural grassland. f_Lij u ^u j*^*Li JL • Journal of Applied Ecology 24: 743-758. from floods could not have a direct influence on their J rr DJ . . Beintema, A.J. (1988): Conservation of grassland bird communities in the Netherland. In: Goriup, P.D. (ed): distribution. What probably shapes their settling patterns are land management, vegetation structure Ecology and Conservation of grassland Birds. ICBP: and food. On Ljubljansko barje, frequently flooded 105-112. areas are not as suitable for agricultural practice as Bra~ko, F. (1986): Naglo upadanje {tevil~nosti zlatovranke others (personal communication with farmers) and Coraeias garrulus v Sloveniji. Acrocephalus 7 (30): 49 - traditional, seldom fertilised, late mowed meadows 5v . 1 « 1 _ 1 _ 1 Bra~ko, F., A. Sovinc, B. [tumberger, P. Trontelj & M. predominate. A late mowing date in particular is . . . v. v ., .... J . , . . , Vogrin (1994): Rde~i seznam ogro`enih ptic gnezdilk known to increase reproductive success in meadow Sl . 11 ,<^7> is? 1 on r Movenije. Acrocephalus O(67): 165-180. birds (Beintema & Muskens 1987), and Chamberlain, D.E. & R.D. Gregory (1999): Coarse and consequently, the breeding density. Corn fields and fine scale habitat associations of breeding Skylarks semi-intensive meadows, which are less friendly to Alauda arvensis in the UK. Bird Study 46: 34-47. 78 Acrocephalus 112.qxd 16. 12. 02 15:29 Page 79 AcROCEPHALus 23 (112): 75 — 79, 2002 Cramp, S., ed. (1998): The complete birds of the western Palearctic on CD-ROM. Oxford University Press. Fuller, R.J., R.D. Gregory, D.W. Gibbons, J.H. Marchant, J.D. Wilson, S.R. Baillie & N. Carter (1995): Population declines and range contractions among lowland farmland birds in Britain. Conservation Biology 9: 1425-1441. Gibbons, D.W., J.B. Reid & R.A. Chapman (1993): The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. Poyser, UK. Grobelnik, V. (2000): Ogla{anje, ocena {tevil~nosti in izbira mikrohabitata koscev (Crex crex, L. 1758) na dveh lokacijah Ljubljanskega barja. Graduation thesis, Univ. Ljubljana. Herkert, J.R. & F.L. Knopf (1998): Research Needs for Grassland Bird Conservation. In: Marzluf, J.M. & R Sallabanks (eds.): Avian Conservation, Island Press, USA: 273 – 283. Kolbezen, M. (1984): Hidrografske zna~ilnosti poplav na Ljubljanskem barju. Geografski zbornik 24: 15-32. Kotarac, M. (1999): Kartiranje habitatnih tipov na Ljubljanskem barju. Poro~ilo za MOP in MOL, 1-11. MartinČiČ, A. (1987): Fragmenti visokega barja na Ljubljanskem barju. Scopolia 14: 1-53. Newton, I. (1998): Bird conservation problems resulting from agricultural intensification in Europe. In: Marzluff, J.M. & R. Sallabanks (eds.): Avian Conservation, Island Press, USA. Schäffer, N. & R.E. Green (2000): The global status of the Corncrake. Conservation review 13: 18-24. Tome, D. (1998): Ali je populacija prib Vanellus vanellus na Ljubljanskem barju pred zlomom? Acrocephalus 19 (90- 91): 130 - 133. Trontelj, P. (2001): Popis kosca Crex crex v Sloveniji leta 1999 ka`e na kratkoro~no stabilno populacijo. Acrocephalus 22 (108): 139-148. SoviNC A., D. Tome & P. Trontelj (1993): Ornitolo{ki atlas Ljubljanskega barja - poro~ilo o poteku popisovanja. Acrocephalus 14 (60): 145- 151. SoviNC, A., D. Tome & P. Trontelj (in prep.): Avifavna Ljubljanskega barja. Tucker, G.M. & M.I. Evans (1997): Habitats for birds in Europe. BirdLife Conservation Series No. 6, Cambridge, UK. Tucker, G.M. & M.E Heath (1994): Birds in Europe, their conservation status. BirdLife Conservation Series No. 3, Cambridge, UK. Willi, P. (1985): Langfristige Bestandestaxirungen im Rheindelta. Egretta 28 (1-2): 1-62. Prispelo / Arrived: 21.6.2002 Sprejeto / Accepted: 23.9.2002 79